As part of devnet 1, we will start with a local docker-compose, to run the full stack.

Unlike op-e2e, this is not as tightly coupled to the current deployment flow, and unblocks runing the interop stack.

For the local devnet we have several options:

1. Extend the current monorepo devnet
2. Reuse some of the current monorepo devnet, but provide a separate entrypoint
3. Copy-paste some of the monorepo devnet
4. Fork of the devnet as a separate repository that deploys the stack

Currently I believe 2 is the best approach, as this can run in CI, while on the same develop branch.

Something similar was implemented here: #10777
However, this targeted an older version of the devnet (before L1 beacon node) and deploys duplicate Superchain contracts to the L1, and misconfigures the addresses of the L2s and their L1 deployments.

We can fork this work, modify the deployment, and then integrate the op-supervisor to unblock interop testing.

The deploy-code should be modified to reuse this for a live long-running devnet.

The primary blocker on getting the devnet running is the genesis-setup. See #11493